Lack of Long-Term Regulatory Framework for AI Raises Concerns

The rapid advancements in artificial intelligence (AI) have been a topic of discussion among experts and the general public. Every few weeks, we hear about new capabilities from the AI community that are pushing the boundaries of what is possible. While this has sparked excitement and curiosity, it also raises concerns about the potential impact on society and jobs.

The simultaneous government announcements of new regulatory actions have received less attention but are fundamentally important. There are over 1,500 bills under consideration in various states, hundreds in Congress, dozens of executive actions from the executive branch, and a diverse set of views among market participants on how to regulate the AI industry.

Initially, I thought that the AI revolution could be governed by the invisible hand of the market. However, with nearly 2,000 proposals being considered, it’s clear that this is not possible. While many proposed policy changes are thoughtful and necessary, they collectively fall short in addressing a crucial issue: establishing a durable, comprehensive future-focused regulatory framework.

National regulatory bodies have traditionally been established after a crisis. The Securities and Exchange Commission (SEC) was created following the 1929 stock market crash, while the Nuclear Regulatory Commission (NRC) was formed after the partial meltdown at Three Mile Island. It’s essential to learn from these examples and not wait for a crisis to occur during this AI revolution.

The establishment of a national regulatory body is not a panacea. A dedicated commission focused on AI impacts would have a dynamic mandate, as it affects various aspects of society. Regulators tend to over-regulate, so continued congressional, executive, judicial, and public oversight is critical. The AI revolution is a global competition, requiring balance between innovation-driven markets and proper regulation.

One aspect of the SEC’s operating model could serve as a partial way forward. In my previous role at Nasdaq, I was surprised by the rules that governed implementing improvements to core exchange technology. We had to submit detailed changes for public comment and subsequent review, which would then be published. This process ensured transparency but also allowed competitors to know exactly what we were planning.

This operating method has contributed to U.S. capital markets being among the best in the world. The SpaceX IPO was only possible on the U.S. market due to these regulations. Similarly, when it comes to major changes to large language models (LLMs), I believe public comments will far exceed those received for other regulatory updates.

I’m advocating for a national regulator for AI without being certain if this is the greater good or the lesser evil. However, I do know that it’s only the beginning. As the AI revolution advances, so must the regulatory apparatus. The SEC of 2026 would be unrecognizable compared to its predecessor in 1934.

Any short-term regulatory efforts will undoubtedly hinder progress, but establishing proper rules for the road is essential for long-term success. It’s crucial that policymakers take proactive steps and establish a national regulatory body with a broad mandate to oversee AI-related opportunities and challenges.